I don't dress my cat in costumes because, without even trying, I know she would
hate that.
But now I can use text-to-image generators like DALL-E2 to imagine what she
would look like in costumes. After all, even if it never saw my cat in
I’m not sure what was more unsettling: my recent failed attempt to train a
neural net called StyleGAN2 to generate screenshots from The Great British
Bakeoff
[https://aiweirdness.com/post/613754259190808576/this-is-not-cozy-ai-attempts-the-great-british]
, or my successful attempt to train StyleGAN2 to generate
Recently I had my first-ever experiment with training a neural net to generate
images, when I trained StyleGAN2 to generate screenshots from The Great British
Bakeoff
[https://aiweirdness.com/post/613754259190808576/this-is-not-cozy-ai-attempts-the-great-british]
. Although recognizable as the baking show, it was a
Last year I trained a neural net to generate new names for kittens
[https://aiweirdness.com/post/162396324452/neural-networks-kittens], by giving
it a list of over 8,000 existing cat names to imitate. Starting from scratch,
with zero knowledge of English or any context for the words and
A new paper from NVIDIA recently made waves with its photorealistic human
portraits. Called StyleGAN [https://github.com/NVlabs/stylegan], the algorithm
had a new training dataset pulled from Flickr, with a wider range of ages and
skin tones than in other portrait datasets. (If your pictures are on Flickr
It was inevitable that the neural network, having named all kinds of internet
datasets, should turn its talents to naming cats.
And what an occasion! The AFK Cat Rescue [https://www.facebook.com/AFKCat/] of
Huntsville, Alabama contacted me because they had an exceptionally adorable
bunch of kittens (plus one
robotlaughs
[https://robotlaughs.tumblr.com/post/160785806733/fun-with-a-neural-net-that-transforms-line]
:
> lewisandquark
[http://lewisandquark.tumblr.com/post/159654519362/fun-with-a-neural-net-that-transforms-line]
:
> So there’s a neural network framework called pix2pix
[https://github.com/phillipi/pix2pix] that can be trained